Zendesk QA scorecard: connect reviews, reopens, and CSAT
A quality program is useful only if it helps the team solve the same problems the customer feels.
Many QA scorecards fail because they measure politeness, formatting, and process compliance while the real issues show up elsewhere: more repeat contact, more reopened tickets, and lower CSAT. Zendesk teams need a QA view that connects review notes to customer outcomes, not a checklist that lives on its own.
This guide explains how to design a Zendesk QA scorecard, what supporting metrics to pair with it, and how to turn review themes into operating changes. For the broader workflow, start with the support metrics dashboard and How to report CSAT in Zendesk.
What a QA scorecard should answer
A strong QA program should help you answer three questions:
- What good support behavior looks like for your team.
- Which quality failures happen most often by issue type, queue, or teammate.
- Whether those failures connect to customer outcomes like reopen rate, repeat contact rate, or CSAT.
If your scorecard cannot do that, it becomes performance theater.
Build the scorecard around observable behaviors
Keep the review rubric short enough that reviewers use it consistently. Four sections are usually enough.
1. Diagnosis quality
Did the agent identify the real issue, ask for the right information, and avoid unnecessary loops?
2. Resolution quality
Was the answer complete, accurate, and likely to prevent repeat contact?
3. Ownership and expectation-setting
Did the customer understand what happens next, who owns it, and when to expect an update?
4. Process discipline
Were the right tags, macros, and internal notes used so the team can learn from the ticket later?
This mix keeps the scorecard tied to both the customer experience and reporting hygiene.
Pair QA with the metrics customers actually feel
QA scores become much more useful when you compare them with adjacent metrics.
Reopens and repeat contact
If tickets with low QA marks also have higher reopened tickets or repeat contact rate, your review process is finding real quality risk.
CSAT
If one issue type or agent group has weaker QA results and lower CSAT, the pattern is probably real enough to prioritize.
Tag-to-resolution time
When the same tags show weak QA and long resolution time, that often points to a product area or troubleshooting workflow that needs better documentation. Use Zendesk tag-to-resolution time report to segment the problem.
First contact resolution and one-touch work
If your team cares about fewer loops, compare QA themes with first contact resolution and replies per ticket. Fast closure is not the same thing as durable resolution.
How to sample tickets well
The easiest way to make QA noisy is to sample only the tickets that are easiest to review.
A better sampling model includes:
- a mix of high-volume workflows and edge cases
- tickets across channels
- tickets with both strong and weak outcomes
- tickets from newer and experienced agents
You do not need to review everything. You do need to avoid reviewing only the most convenient tickets.
How to report QA themes in Zendesk
Zendesk does not need to become a dedicated QA platform for this to work. A practical QA reporting setup usually includes:
- A review rubric stored consistently.
- Theme tags or categories for common failure types.
- A monthly trend of QA findings by team, issue type, or workflow.
- A comparison view against reopens, CSAT, and repeat contact.
The point is to move from isolated ticket review to pattern detection.
What to do with weak QA signals
When one theme keeps appearing, connect it to an operating fix.
- Weak diagnosis quality: improve troubleshooting paths or macros.
- Weak expectation-setting: update communication templates and follow-up norms.
- Weak ownership: clarify handoff rules and escalation paths.
- Weak process discipline: simplify required tags or notes so compliance is realistic.
This is where QA becomes useful. It stops being a score and starts becoming a workflow improvement engine.
Common mistakes
- Overweighting tone and underweighting resolution quality.
- Reviewing too many categories for reviewers to apply consistently.
- Keeping QA separate from customer outcomes.
- Using QA only for coaching individuals when the pattern is systemic.
FAQ
How many QA criteria should we use?
Use the fewest criteria that still capture diagnosis, resolution quality, ownership, and process discipline.
Should QA be tied to performance reviews?
Carefully. QA is most valuable when it drives learning and process improvement, not just scorekeeping.
What if QA looks fine but customers still reopen tickets?
That usually means the scorecard is missing a behavior that matters, or reviewers are not measuring durable resolution strongly enough.
Connect QA reviews to the support outcomes that matter - start free